I've always thought iron lungs are fascinating.
My fascination with them began with T. Herman Zweibel's column in The Onion, a spectacular example of which you can read here: https://www.theonion.com/i-am-leaving-the-bloated-corpse-of-journalism-behind-fo-1826907443 (yes I know this is a made up character in a satirical publication, but it's more fun to think it's real)
Here's an example of Zweibel writing about his iron lung: https://www.theonion.com/the-peace-of-the-womb-1819583764
Anyway, there are, according to Gizmodo, perhaps a half-dozen iron lungs in use in the US: https://gizmodo.com/the-last-of-the-iron-lungs-1819079169
This is mostly a good thing. Hulking iron lungs have been replaced by positive pressure ventilators like CPAPs. Apparently they're mostly as good. Gizmodo:
When Lillard is outside of the tank, she can breathe using a positive-pressure ventilator, a smaller device that pushes air into her lungs. But that instrument doesn’t provide the same relief as when she puts her entire body into the 640-pound, 7-and-a-half-foot-long apparatus. Plus forcing air into the lungs can cause inflammation or damage the air sacs. When she’s sick, she can only heal if she spends full days in the iron lung.
This will sound familiar to readers of The Innovator's Dilemma.
And this part of the article will sound familiar to folks who work with legacy systems:
Her iron lung has portholes and windows on the side; a pressure gauge at the top. The machine is actually cobbled together from two iron lungs. One, the March of Dimes gave her when she was a child. The other, she bought from someone in Utah, after she haggled him down from $25,000 to $8,000. The body has also been modified over the years. Her grandfather invented a motorized pulley system that closes the bed tray into the tank after she climbs in. He also replaced the brushed aluminum mirror above the neck slot with a real mirror so that she could have a clear view to the rest of the room when she’s locked in the canister. A local engineer used a motor from an old voter registration device to build a mechanism that tightens the collar around her neck after she slips her head through the portal. The fan belts and half-horsepower motor have been replaced about ten times.
So it goes with technology. Platforms get supplanted with what's newer, even if it's not as "good" however it is that you define good.
I think I've pretty well documented the constraints and shortcomings of platform specialization. Any situation where your core skills become commoditized in 5 to 7 years is, to me, a young person's game. You can make a decent living with commoditized skills, but without licensing and other barriers to entry, fully commoditized tech skills can leave you charging less per hour than a plumber or car mechanic.
If you're committed to the platform specialization route, the answer is to cultivate a genuine interest or sympathy to the business impact of the technology you're focused on.
I recently had a super interesting conversation with Corey Quinn, who has done just that. I'm gonna publish this conversation on my podcast in a few weeks, and so if you aren't subscribed to the podcast now might be a good time to do so: http://consultingpipelinepodcast.com